Our goal is to define the mechanism by which the androgen receptor (AR) initiates prostate cancer growth. We hypothesize that prostate cancer growth is mediated through androgen receptor regulation of specific cell cycle regulatory molecules. Our specific aims delineate the mechanism by which androgen stimulation affects cell cycle regulatory molecules in prostate cancer cells, and to determine if the expression of cell cycle regulatory molecules affected by AR in cell line LNCaP correlates with tumor differentiation and disease progression in human prostate cancer xenografts and archival tumors. Delineation of the mechanism of AR regulation the of cell cycle offers the potential to identify novel therapeutic targets in prostate cancer. We have observed that androgen stimulation of an androgen responsive prostate cancer cell line LNCaP results in increase growth, increased expression of cyclin A and cyclin dependent kinases (cdk's) 1 and 2, and decreased expression of the retinoblastoma protein (Rb). We have shown by mRNA analysis that cdk2 induction and Rb suppression occur post-transcriptionally, and we will determine the nature of cdk 1 and cyclin A induction. We will determine the mechanism of post-transcriptional regulation of Rb and cdk 2 by 35S methionine pulse chase experiments. If degradation is observed, based on previous reports of ubiquitin medicated degradation of cell cycle regulatory molecules, we will determine if immunopurified cdk 2 and Rb in LNCaP are ubiquitinated in the absence or presence of androgen by ubiquitin immunoblotting. In vitro degradation assays using LNCaP cell extracts, with or without endogenous 26S proteasome removed, will be performed to determine if ubiquitin mediated degradation of either protein occurs in these cells. The relationship of cyclin A/cdk 1 or 2 complexes to Rb degradation and cell growth will be determined by immune complex kinase assay of Rb substrate and 2-D tryptic phosphopeptide mapping of Rb. The pattern of cyclin A/cdk 1 or 2 dependent Rb phosphorylation will be defined, and the ability of this kinase activity to target Rb for ubiquitination will be determined by in vitro degradation assays of phosphorylated Rb. Early response genes of AR involved in growth will be identified by mRNA representational difference analysis and characterized by selective hybridization with cDNA's strongly induced by androgen treatment of LNCaP. The expression of cyclin A, cdk 1, cdk 2, and Rb will be measured in xenograft models of human prostate cancer by immunoblotting, and in archival paraffin embedded tumors by immunohistochemistry and correlated with tumor growth, grade, stage, and outcome.